A PHASE II TRIAL OF REDUCED DOSE BRENTUXIMAB VEDOTIN FOR CUTANEOUS T‐CELL LYMPHOMAS

Abstract

Introduction: Mycosis fungoides (MF) is the most common cutaneous T cell lymphoma. Brentuximab vedotin (BV) is a highly effective therapy for MF with an overall response rate (ORR) of 65% in the phase III ALCANZA trial (Prince et al, Lancet 2017). However, 44/66 (67%) of patients on BV developed peripheral neuropathy (PN) (G2, n = 21; G3, n = 6), limiting treatment duration (median, 12 cycles). We hypothesized that lower dose BV would produce acceptable efficacy with reduced PN for the treatment of MF. Method(s): This is a multicenter phase II study. We treated BV-naive patients with MF (n = 19) in a Simon two-stage design with expected ORR 35% and desired ORR of 65%. Patients were treated with BV at 0.9 mg/kg. If <11 responses were seen at the end of the second stage, the Simon 2 stage design restarted with BV at 1.2 mg/kg. Responses are graded every 4 cycles. Global response is graded using the International Society for Cutaneous Lymphomas consensus criteria (Olsen et al, JCO 2011). PN is graded using Common Terminology Criteria for Adverse Events v. 4.0 (CTCAE) and the Total Neuropathy Score Clinical (TNSc). Health Related Quality of Life (HRQoL) is self-reported by patients using the FACT-Lym, FACTGOG-NTX and Skindex-29 questionnaires. Result(s): 19 patients were treated in Cohort 1 with BV at 0.9 mg/kg, with ORR of 42% (8 partial responses (PR), 0 complete responses (CR)) (Fig 1A). Median PFS was 21.9 months (95% CI, 4.8-NR) and median duration of response was 21.1 mon (95% CI, 5.9-NR). Median time to best response was 2.5 months (range, 2.1-7.8 mon), or 3 cycles (range, 3-11 cycles) (Fig 1A). 8/19 (42%; 95% CI, 20-66%) patients developed PN. 5/8 (63%) developed grade 1 (G1) PN at a median of 5 cycles (range, 3.5-11); 3/8 (37%) patients developed G2 PN at a median of 18 cycles (range, 13-19). Median time of PN recovery from G2 to G1 is 1.6 mon; median recovery from PN G1 to resolution is 6.1 mon. Additional BV-related adverse events seen in >10% of patients included increased ALT (G1, n = 4), increased AST (G1, n = 3; G2, n = 1), diarrhea (G1, n = 3), fatigue (G1, n = 2; G2, n = 1); infusion-related reaction (G1, n = 3; G2, n = 1), localized edema (G1, n = 2), maculopapular rash (G1, n = 1; G2, n = 1; G3, n = 1), pain in extremity (G2, n = 2), pruritus (G1, n = 2; G3, n = 1) and nausea (G1, n = 2). Conclusion(s): Treatment with BV at 0.9 mg/kg resulted in an ORR of 42% (95% CI, 20-66%) which did not meet the primary efficacy endpoint. However, we observed a lower rate of >=G2 PN than reported with full dose therapy in ALCANZA. We are now enrolling BVnaive patients in Cohort 1 for treatment with BV at 1.2 mg/kg. We are also enrolling patients in an exploratory cohort of patients previously treated with BV, and a second exploratory cohort of patients with LyP. We continue to collect correlative data on HRQoL, PN and the tumor microenvironment.